Oops! The input is malformed!
Originally published 19 January 2006
When is the data warehouse finished? This is a frequently asked question. The answer is that the data warehouse is finished when the corporation stops changing. Only when there are no more new customers, products, competitors, or reasons to cut expenses, can the data warehouse be finished. Thus, the data warehouse is completed when the company goes out of business. In other words, as long as the corporation is viable, the data warehouse is evolving.
In fact, the data warehouse is always being built and improved. This is true as long as the company is in business, business conditions are changing and there are marketplaces to explore.
This means that the data warehouse is an ongoing effort. Of course, the effort required to build the data warehouse changes. Obviously, building the first few iterations of the data warehouse requires much more effort than adding to it in the first few years after it was built. The initial construction of the data warehouse requires considerable resources, regardless of your chosen technology and methodology.
But continuing and completing a data warehouse are two completely different issues.
When can I start to achieve usable results from my data warehouse? This is a related question that is frequently asked about data warehousing. While you cannot get project results for many IT issues until after the fact, this is fortunately not true for data warehousing.
Once you have completed your first iteration and the initial data has been loaded into the data warehouse, you can get results. Ideally, data warehouses should be built in iterations. All iterations cause some data to be populated into the data warehouse. After data is first populated into the warehouse, the data warehouse can become usable. Although the data warehouse is not fully populated when the first iteration is complete, it can be used.
When am I finished building my data warehouse? When can I begin obtaining results from my data warehouse? These two questions sound the same. The answers to each one, however, are very different.
The completion and usability of data is important since there used to be a “hold your breath” mentality in our industry. Because organizations could not get any tangible results until the project was finished, many simply had to wait. Fortunately, this approach isn’t needed for the data warehouse. Clearly, the data warehouse provides a tangible data source for your ongoing business. It can also capture necessary data for your business to move, and ultimately thrive.
SOURCE: The Evolutionary Data Warehouse
Recent articles by Bill Inmon